Background

Dyspnea is a hallmark symptom of idiopathic pulmonary fibrosis (IPF), and dyspnea induced physical activity limitation is a prominent driver of quality of life impairment among IPF patients.

Methods

We examined response data for the 21 physical activity items (the first 21 of 24) from the University of California San Diego Shortness of Breath Questionnaire (UCSD) collected at baseline in a recently conducted IPF trial. We used Rasch analysis and hypothesis testing with conventional statistical methodology to achieve three objectives: 1) to examine the items to identify the one characteristic that distinguishes one from another; 2) to asses these items for their ability to measure dyspnea severity in IPF; 3) to use the items to develop a dyspnea ruler.

Results

The sample comprised 178 subjects. The 21 items fit the Rasch model. There was very strong correlation between Rasch item severity and their metabolic equivalents (METS) values (r = -0.86, p < 0.0001). With the sample stratified on scores from the 21 items, there were significant between group differences in FVC%, DLCO% and distance walked during the six-minute walk test. The dyspnea ruler can be used to put dyspnea levels in a more easily understood clinical context.

Conclusions

The first 21 items from the UCSD compose a unidimensional dyspnea-with-activity scale and are both sensibly ordered and distinguished from each other by their METS values. These 21 items can be used confidently to formulate clinically-relevant inferences about IPF patients and should be considered for use as a meaningful endpoint in IPF research.